While significant effort has been put into protecting users from harmful emails, some of the most devastating attacks continue to be launched via voice-to-voice communications. Elaborate fraud rings have been built around cold calling people and making claims of back taxes owed or lottery winnings. These attacks have cheated people out of hundreds of millions of dollars. Many of these attacks are not targeted attempts with a single victim, but instead, masses of similar or identical scam calls sent to arbitrary recipients. While many recipients see through the scam and immediately rebuff the caller or hang up, some recipients may be unaware of the scam and may not take the initiative to do the relevant research before sending large sums of money to the caller.
Unfortunately, most traditional systems for preventing malicious communication of this sort are directed at emails. While some traditional systems for identifying suspicious calls exist, these systems often rely on vigilant recipients to manually report the attempted scam call. Most recipients of a suspicious call may not follow through with a manual reporting process, leaving the system with little data to use to combat future suspicious calls. Some traditional systems rely on identifying features such as the caller's number, which can be easily changed with a variety of caller identification spoofing applications. The instant disclosure, therefore, identifies and addresses a need for systems and methods for detecting suspicious voice calls.